Skip to main content

Glossary

B

Butterfly Plot

Shows the ERP of all electrodes on a single figure. See Ch 9

C

Cognitive Electrophysiology

The study of how cognitive functions are implemented by the electrical activity produces by neuron populations. A spectrum ranging from cognitive studies (psychology)) to Electrophysiology (neuroscience)

Convolution

When a dot product is done upon a segment of a data vector using a reversed kernel vector, essentially a time-domain extension of dot product. Important basis for many statistical analysis. See Ch 10

Convolution Theory

Sates that Convolution in the time domain is equivalent to (per point) multiplication in the frequency domain. Is used to make computation faster. See Ch 11

Cross-Covariance

Similar to a Convolution, but the kernel is not flipped. See Ch 10

D

Dot Product

A linear measure of similarity between 2 vectors. The basis of convolution. See Ch 10

E

Electrocorticogram (ECoG) / Intracranial EEG

Measurement of the electric potential in the extracellular space around neurons. Invasive as it is done on the brain directly using surface electrodes

Electroencephalography (EEG)

The measurement of electrical fields of the brain externally via the use of numerous electrodes

  • Pros:
    • High Temporal Resolution
    • Directly measures neuronal population activity
    • Measures across time, space / location, frequency, and power / strength
  • Cons:
    • Not good for precise location analysis
    • Not good for trials longer than a few seconds

See Ch 2

Event Marker

On a separate channel of the EEG, an Event marker is a square wave that is sent to denote an event (such as a button press). See Ch 6

Small voltages from brain structures in response to specific events. Regarded as Time Locked activity. See Ch 2, Ch 5, Ch 9

F

Flicker Effect

The tendency for brain activity to sync frequency with external factors (such as light flashing). Known as SSVEP (Steady-State Visual Evoked Potential) with the visual system and SSAEP (Steady-State Auditory Evoked Potential). See Ch 9

Fast Fourier Transform

A more complex implementation of the Fourier Transform that is much faster computationally, is essentially always used. See Ch 11

Fourier Transform

Converts time-series data into a 3D output showing the frequency, power, and phase. See Ch 11

Functional Magnetic Resonance Imaging (fMRI)

An extension of MRI, in which MRI machines are used to scan the brain for changes in blood oxidation, which implies neuronal activity. This is sometimes also referred to as BOLD (Blood oxygenation level dependent) imaging. See Ch 2

G

Global Field Power (GFP)

The standard deviation of activity of all electrodes / how "different" activity is at different electrodes. Not to be confused with the average power of all electrodes. See Ch 9

Global Map Dissimilarity 

Also known as the temporal difference, refers to the amount of change over all electrodes at a given time point. Essentially the index of change of the Global Field Power.

I

Independent Component Analysis

The act of separate EEG data into pieces to identify independent sources of noise and signal. See Ch 8

L

Local Field Potential Recording (LFP)

Measurement of the electric potential in the extracellular space around neurons. Invasive as it is done on the brain directly using depth electrodes

M

Magnetoencephalography (MEG)

The measurement of magnetic fields of the brain externally via the use of numerous magnetic sensors in a magnetically isolated environment. See Ch 2

MATLAB

A stupid programming language where indexing starts at 1. Commonly used in research. Lots of EEG packages are designed for it. Look into Octave for similar free functionality

Microstates

Sub-Second intervals in which brain states (electrical activity) is quasi-stable (relatively constant). Often correlated to cognitive tasks such as memory recall. See Ch 9

N

Neurotransmitter

Chemical signaling molecules that neurons use to communicate with each other. As they are not electrical, they are not directly detectable by EEG. See Ch 5

Nyquists Theorem

States that one would need to sample at least twice per period of a signal to accurately deduce the source wave. See Ch 6

O

Oscillation

A periodic change between 2 values. In the context of neuroscience, cognitive processes are thought to be largely encoded in neural oscillations around the brain. See Ch 5

P

Preprocessing

Refers to processing of the raw data from the EEG trials before actually running any analysis. Generally includes removing noise from data sets. Note that while preprocessing data with some noise can improve the end result, no amount of preprocessing can save extremely noisy data. See Ch 7

S

Stationarity

Refers to how stationary (or "well-behaved") a signal is, in other words how stable the signal is. See Ch 11

T

Topographical Map

Shows activity over a physical area of electrical activity. See Ch 9

Back to top